I am currently working/starting on a second design of my dual extruder setup. it is a ‘standard’ setup with 2 indivudial motors nothing weird about it. But i am trying to make it as compact as possible.
Having the air go through the fins and then through a narrow space upwards is going to seriously restrict airflow. For that reason, I’d be very leery about using it.
The second option “push-pull” is going to move heated air across another heater - I just pulled out some of my old thermodynamics textbooks and I think the loss in efficiency will be much less than 1% in a “standard” room, 20C at 1 bar air temperature and assuming the temperature of the air passing through the first heatsink is not raised by more than 5C. (The isochoric heat capacity will go from 0.7172 kJ/(kgK) at 20C to 0.7175 kJ/(kgK) at 25C).
Please note that it has been about 20 years since I last had to do this type of work (which was air cooling for servers) so take the numbers with a grain of salt - the conclusion is probably accurate but the numbers might not be totally accurate or appropriate for your location (ie if you were at a high elevation).
You might want to consider a third option - that is to use your “push-push” but have the heated air vent forward and backward in the toolhead rather than up. This may affect your esthetics for your toolhead and may make placing your Z axis sensor more difficult. The advantage of this is that the exiting air has minimal restrictions.
The “best” solution would probably be water cooling - no difference between the cooling going to the two extruders, minimal toolhead volume and no fans or vents to get in the way of things.
Personally, I would probably go with the “push-pull” option as it is probably “good enough” and won’t require a lot of work.
My concern is that the push-pull mechanism effectively creates a short circuit. It would only work efficiently if there is no leak, such as in a sealed pipe. Otherwise, the air will simply take the easiest route, which is definitely not through two heat sinks.
Having the air go through the fins and then through a narrow space upwards is going to seriously restrict airflow. For that reason, I’d be very leery about using it.
Yes, i had a feeling that would happen but now it is confirmed.
You might want to consider a third option - that is to use your “push-push” but have the heated air vent forward and backward in the toolhead rather than up.
I can’t do that due to the fact that there a also ducted fans (for the nozzle) on the front and back of the toolhead, so that will be difficult…
My concern is that the push-pull mechanism effectively creates a short circuit.
well, given the fact that the top, front and back are closed and i can minimalize the opening on the bottom it would work.
and when the ducted fans are on, they also create a kind of wall of air on the bottom with a higher pressure then the noctua fans.
The “best” solution would probably be water cooling - no difference between the cooling going to the two extruders, minimal toolhead volume and no fans or vents to get in the way of things.
well yeah, but water cooling on a 3d-printer is pretty overkill in my opinion. And i also do not have the room, tools, patience and experience to start with water cooling.
Let us know what you decide.
This was my go-to plan, but i just came here to ask if this is my best option. because other people have different experiences and knowledge and i can learn a lot from that.
But i think i will stick with the push-pull mechanism and try to enclose it as much as possible.
the finest i can get the distance between the 2 heatsinks is 3.70mm.
Is this Ok? What i am thinking atm is that it might interrupt the airflow a bit. But i don’t have the knowledge/experience to indicate how much it will be interrupted.
Ofcourse leaving the rest of the factor out of sight and assuming that it is fully enclodes.
Are you worried about a 3.7mm space between the heatsinks?
You should be fine, just as @Sineos noted that it is sealed all the way around and there aren’t any significant air leaks.
I would think that you should be able to dummy up something pretty quickly (even with something like cardboard) and test it out, using an additional thermistor as a temperature probe.
When doing these experiments you should start at something like 50C, then go to 100C, 150C, watching both the temperature and PWM of the two extruder/heatsink assemblies to make sure nothing unexpected is happening.
I’m saying to monitor the PWM because that is an indication of how much heat is going into the extruders and, at the same temperature, it should be basically identical for each of them set at the same temperature. If they are significantly different, then you’ve got a problem.
The klippy.log analysis tool that @Sineos created is actually ideally suited for this.
